A precision 32 keV angular-selective photoelectron source for calibration measurements at the KATRIN experiment

Abstract

The Karlsruhe Tritium Neutrino (KATRIN) experiment measures the neutrino mass from a precise measurement of the endpoint region of the kinematic tritium beta-decay spectrum by using a spectrometer combining magnetic adiabatic collimation and electrostatic filtering (MAC-E filter). For calibration purposes, KATRIN uses a monoenergetic angular-selective photoelectron source. We present an upgrade of this source, which was installed in the KATRIN beamline in February 2022. The source allows for a wide range of accessible electron energies up to 32 keV and a variation of the angle with regard to the magnetic field. These features are used for precise measurements of electron scattering effects off tritium molecules in KATRIN's gaseous tritium source, for investigations of angular-dependent backscattering for example at KATRIN's focal-plane detector, and for studies on adiabatic transport in the main spectrometer.

Figures (9)

• Figure 1: Schematic working principle of the photoelectron source. UV light is guided via an optical fibre towards the photocathode, where low-energy electrons are emitted. The cathode is placed on the backplate and put at a negative high voltage, so the electrons are accelerated by the electric field $\vec{E}$ between back- and frontplate. The electron angle can be altered by adjusting the plate angle $\alpha_\text{P}$ with regard to the magnetic field $\vec{B}$.
• Figure 2: Illustration of the electron-source setup at the KATRIN Rear Section. The backplate voltage is defined by the voltage of the HV cage and an offset voltage $U_\text{off}$ within $\pm 500V$. The frontplate voltage is provided by an additional high voltage $U_\text{acc}$. The electric potential is stepwise reduced to ground via three post-acceleration electrodes connected to a HV divider. The acceleration takes place in the first 0.5m of the Rear Section with a total length of around 5m. The electron beam is steered by several magnetic dipolesteering coils (not shown) and a set of dipole electrodes to pass the Rear-Wall hole and the chicane towards the beamline. The chicane prevents damage from the electron source by avoiding that neutral molecules from the tritium source can directly travel towards the sensitive photocathode of the electron source. One of the dipole electrodes can be pulsed via a MOSFET half bridge for background reduction, see sec. \ref{['sec:background']}. The guiding magnetic field in the Rear Section is provided by several solenoids.
• Figure 3: CAD drawing (left) and picture (right) of the upgraded photoelectron source. The source has a total length of 33cm (including the motor), and is hosted on a CF160 flange.
• Figure 4: Transmission function of the photoelectron source measured at a standard energy of 18.8keV (top) and an enhanced energy of 32.2keV (bottom). Both graphs show the electron rate measured at the detector as function of electron energy $E_\text{kin} = qU_\text{back}$, with the spectrometer set to the retarding voltage $qU$ indicated by the legend. The dashed lines show the $1\sigma$ width of the functions. The maximum visible rate is around 25% of the total electron rate, because the measurements were taken while the beamtube was filled with tritium gas, such that $\approx 75%$ of the electrons lost energy due to inelastic scattering off tritium molecules and are not observed in the shown energy range.
• Figure 5: Transmission functions measured at different UV-light wavelengths. Shown is the normalized measured electron rate at the detector as function of the surplus energy $\Delta E_\text{s} = E_\perp - qU$, taken at a magnetic field in the analyzing plane of $B_\text{ana} = 1G$. The energy distribution gets broader for higher photon energies corresponding to lower wavelengths. The energy spread can be approximated as the width $\sigma$ of an error function fitted to the data, the obtained values for the three measurements are also shown in the figure.